Background: With evidence of sustained transmission in more than 190 countries, coronavirus disease 2019 (COVID-19) has been declared a global pandemic. As such, data are urgently needed about risk factors associated with clinical outcomes. Methods: A retrospective chart review of 323 hospitalized patients with COVID-19 in Wuhan was conducted. Patients were classified into three disease severity groups (non-severe, severe, and critical), based on their initial clinical presentation. Clinical outcomes were designated as favorable and unfavorable, based on disease progression and response to treatments. Logistic regression models were performed to identify factors associated with clinical outcomes, and log-rank test was conducted for the association with clinical progression. Results: Current standard treatments did not show significant improvement on patient outcomes in the study. By univariate logistic regression model, 27 risk factors were significantly associated with clinical outcomes. Further, multivariate regression indicated that age over 65 years, smoking, critical disease status, diabetes, high hypersensitive troponin I (>0.04 pg/mL), leukocytosis (>10 × 10 ⁹ /L) and neutrophilia (>75 × 10 ⁹ /L) predicted unfavorable clinical outcomes. By contrast, the use of hypnotics was significantly associated with favorable outcomes. Survival analysis also confirmed that patients receiving hypnotics had significantly better survival. Conclusions: To our knowledge, this is the first indication that hypnotics could be an effective ancillary treatment for COVID-19. We also found that novel risk factors, such as higher hypersensitive troponin I, predicted poor clinical outcomes. Overall, our study provides useful data to guide early clinical decision making to reduce mortality and improve clinical outcomes of COVID-19. (Funded by the Natural Science Foundation of Hubei Province ZRMS2019000029 and the Top Youth Talent Program in Hubei Province.)

Background. Type 2 diabetes (T2D) as a worldwide chronic disease combined with the COVID-19 pandemic prompts the need for improving the management of hospitalized COVID-19 patients with preexisting T2D to reduce complications and the risk of death. This study aimed to identify clinical factors associated with COVID-19 outcomes specifically targeted at T2D patients and build an individualized risk prediction nomogram for risk stratification and early clinical intervention to reduce mortality. Methods. In this retrospective study, the clinical characteristics of 382 confirmed COVID-19 patients, consisting of 108 with and 274 without preexisting T2D, from January 8 to March 7, 2020, in Tianyou Hospital in Wuhan, China, were collected and analyzed. Univariate and multivariate Cox regression models were performed to identify specific clinical factors associated with mortality of COVID-19 patients with T2D. An individualized risk prediction nomogram was developed and evaluated by discrimination and calibration. Results. Nearly 15% (16/108) of hospitalized COVID-19 patients with T2D died. Twelve risk factors predictive of mortality were identified. Older age (HR = 1.076, 95% CI = 1.014–1.143, ), elevated glucose level (HR = 1.153, 95% CI = 1.038–1.28, ), increased serum amyloid A (SAA) (HR = 1.007, 95% CI = 1.001–1.014, ), diabetes treatment with only oral diabetes medication (HR = 0.152, 95%CI = 0.032–0.73, ), and oral medication plus insulin (HR = 0.095, 95%CI = 0.019–0.462, ) were independent prognostic factors. A nomogram based on these prognostic factors was built for early prediction of 7-day, 14-day, and 21-day survival of diabetes patients. High concordance index (C-index) was achieved, and the calibration curves showed the model had good prediction ability within three weeks of COVID-19 onset. Conclusions. By incorporating specific prognostic factors, this study provided a user-friendly graphical risk prediction tool for clinicians to quickly identify high-risk T2D patients hospitalized for COVID-19.

The identification of robust biomarkers and tractable drug targets is the foundation upon which modern precision pharmacology is built. For decades, this enterprise has been driven by hypothesis-led, mechanism-anchored experimentation that combines molecular biology, classical pharmacology, and increasingly powerful omics technologies. These traditional approaches have produced an enduring catalog of clinically meaningful markers and targets-from serum proteins and tissue biomarkers to actionable oncogenic drivers-and continue to deliver the mechanistic depth required for translational confidence (FDA-NIH Biomarker Working Group, 2016). At the same time, the scale, dimensionality, and heterogeneity of contemporary biomedical data have outpaced what conventional analytical pipelines can comfortably accommodate, exposing meaningful gaps in reproducibility, integration across data modalities, and prioritization of candidates for downstream validation.Over the past five years, artificial intelligence (AI), including machine learning (ML) and deep learning, has emerged as a complementary force in this space. AI methods excel at extracting compact signatures from high-dimensional data, integrating heterogeneous evidence layers, and learning representations that can be transferred across diseases (Topol, 2019;Vamathevan et al., 2019). They also enable rapid drug-target and drug-response inference at a scope that hypothesis-driven studies cannot achieve in isolation. Yet AI is not a substitute for biology. Concerns about overfitting in small cohorts, fragile generalization across centers, model opacity, and the gap between in-silico predictions and clinical utility remain genuine impediments to translation. The most credible path forward is, therefore, neither purely traditional nor purely computational, but iterative: AI prioritizes hypotheses from large datasets; classical experimental and clinical pipelines validate, refine, and contextualize them. This Research Topic, Advances in Biomarkers and Drug Targets: Harnessing Traditional and AI Approaches for Novel Therapeutic Mechanisms, was launched with this iterative vision in mind.We sought contributions that (i) deploy traditional omics or molecular biology to define new biomarkers and targets, (ii) apply AI to extend or accelerate such discoveries, or (iii) explicitly integrate both. The collection that has emerged spans oncology and major chronic noncommunicable diseases, includes original research, mini reviews, and reviews, and showcasesrather than catalogs-the methodological logic that is reshaping pharmacological discovery. In this Editorial, we synthesize the principal themes, highlight representative contributions, and outline the open questions that the field must address as it moves from data-rich discovery to clinically robust translation.Oncology has been the most active domain in the Research Topic, reflecting both the urgency of unmet clinical needs and the comparative availability of large molecular datasets. Several contributions illustrate complementary modes by which traditional and AI strategies can be combined to refine biomarker discovery, target prioritization, and treatment-response prediction across tumor types.In non-small cell lung cancer (NSCLC), Qin et al. interrogated the predictive value of mitochondria-derived RNAs (mtRNAs) for chemotherapy response by combining classical machine-learning algorithms with the BiomedGPT framework. Their analysis identified mtRNA features associated with differential chemotherapy benefit and underscored how AI-enabled analytics can extend the utility of biomarker classes that have previously been studied mainly in a diagnostic-rather than predictive-context. The work provides a methodologically explicit example of how generative biomedical models can be wrapped around conventional ML pipelines to refine therapeutic decision-making in NSCLC, while also acknowledging the standard caveats around cohort size and external validation.Wan et al. focused on thyroid cancer, where the absence of robust molecular biomarkers continues to limit risk stratification and target selection. Their study integrated bulk-transcriptomic meta-analysis with ML-driven feature selection and single-cell RNA-sequencing of more than 50,000 thyroid-cancer cells, then validated key findings in TPC-1 and BHT101 cell lines using miR-6756-5p overexpression and CRISPRi-mediated knockdown, including xenograft experiments. The convergence of these layers yielded a four-feature panel-BID, MIR6756, ITM2A, and TGM2-with strong discriminative performance in training and independent validation sets, and a defined therapeutic axis through miR-6756-5p. This is, in many ways, the prototypical workflow this Research Topic was designed to encourage: AI is used to compress high-dimensional data into a small set of testable candidates, while traditional functional assays establish biological plausibility and mechanism.rather than supplant traditional biomarkers? Focusing on hepatocellular carcinoma (HCC) recovery, the authors review how AI-enabled analyses can be embedded around long-standing markers such as AFP and PIVKA-II, alongside radiomics-derived imaging features, to produce calibrated, interval-specific relapse-risk estimates that support risk-stratified surveillance. They argue-correctly, in our view-that AI-enabled validation in HCC recovery should proceed as a structured pathway that links analytical validity, clinical validity, and clinical utility while preserving the stated intended use. Their treatment of DECIDE-AI for early-stage live evaluation and of SPIRIT-AI/CONSORT-AI for randomized designs offers a usable scaffold for translational rigor.Finally, in the setting of osteosarcoma, Gao and Wu provide a mini review that articulates an explicit operational definition of "integration": AI-guided analyses narrow and organize candidates, while traditional experiments provide mechanistic support and functional...

[This corrects the article DOI: 10.1155/2022/9322332.].